Paul Krugman on interest rates and gold: “Mr. Magoo, you’ve done it again”
Once again, Paul Krugman manages to stumble Mr. Magoo-like to his analytical destination through a series of comical errors.
Krugman’s argument on gold and deflation is actually an argument on gold and depressions. Krugman begins by explaining that rising gold price has been popularly linked to the prospect of inflation created by the monetary policies of the Federal Reserve Bank. He has ignored this argument, because he thinks Fed policy is far too restrictive to create inflation — deflation is his worry. He has, in fact, brushed rising gold prices aside as something caused by gold-bugs and the like — until now.
Now, he thinks, he can explain why rising gold price may actually be an expected outcome of a deflation, not inflation. I know Krugman’s argument here is flawed, but coincidentally on the right side of the relation: rising gold price implies the economy is experiencing a depression but this real contraction of economic activity does not necessarily lead to a general fall of prices — the deflation Krugman thinks he can explain. So, I want to examine his argument to locate the fallacy in it.
In the model Krugman is using, gold is an ordinary commodity like oil or coal; i.e., without any significant monetary properties. Gold is used primarily for its industrial applications:
Imagine that there’s a fixed stock of gold available right now, and that over time this stock gradually disappears into real-world uses like dentistry. (Yes, gold gets mined, and there’s a more or less perpetual demand for gold that just sits there; never mind for now).
At this point, we need to make explicit what Krugman wants to dismiss in the set up of his argument: First, he is dismissing what is undeniably the most important use of gold: its use as money, as measure of value and as standard of prices. The use of gold as a way to store value — as gold “that just sits there” — does not consume the gold; it simply sits in a bank vault or some other storage facility and is rarely if ever moved, except to be transferred to the ownership of another person. What makes gold ideal for this is that it has a shelf-life that is unlimited — because it does not corrode or otherwise decompose. Even as standard of price gold does not necessarily get consumed. If it is used as currency it may be eroded during the course of circulation. But if it is not directly used as currency, this is not true — again, it simply sits in a bank vault until it is exchanged for paper tokens of itself.
Second, Krugman wants us to ignore the fact that the existing stock of gold is constantly being added to by production of new gold from sources deep in the earth. Most of this new gold also does not enter into production, but is used for its principal purpose as money — as a store of value (savings). Production of gold has to be important in any explanation because of a unique characteristic this gold production has: the production of gold does not appear to be significantly affected by the laws of supply and demand. While the price of gold may rise or fall, the amount of gold produced manages to remain in a very narrow band; rarely, if ever falling out of this narrow limit — e.g. between 2001 and 2010 production ranged between 2400 to 2650 tons per year, while prices quadrupled. As a commodity, gold behaves very curiously in a non-commodity fashion
These two objections are enough to raise serious questions about Paul’s entire model, but, for the moment, we will set them aside and continue to examine Paul’s argument:
The rate at which gold disappears into teeth — the flow demand for gold, in tons per year — depends on its real price
We have a fixed stock of gold that is gradually being consumed by various uses in production. Krugman argues that the rate this stock of gold is consumed will depend on its “real” price. What is the “real” price of gold, and how does this differ from the nominal currency price of gold? Krugman does not tell me. He simply throws the term out there and expects me to figure it out for myself. Since, I can only price gold in an existing currency, I assume by “real” price, Krugman means its currency, e.g. dollar, price. We will see why my assumption is not be correct — gold, it turns out, does not have a price, “real” or otherwise. For now, let’s continue:
Crucially, at least for tractability, there is a “choke price” — a price at which flow demand goes to zero. As we’ll see next, this price helps tie down the price path.
Krugman is arguing there is a price at which the “flow demand” (the money demand for a good over time) for gold in the market goes to zero. He slips this assumption into his argument without discussing it, but I am forced to wonder how he arrives at this statement. Certainly, for use as an ordinary commodity, as a commodity used in industrial processes, we can assume there is a point at which the price of gold might become prohibitive. But, as money — as store of value, or as the standard of prices — is there any evidence that gold has a price point at which demand for its goes to zero? Well, no and yes. One of the paradoxes of gold is that demand tends to increase along with the price. Here is just one example taken from a gold-bug (he even calls himself “Mr. Gold”) doing research on China’s demand for gold:
When at the beginning of this century I studied the elasticity of gold demand to incomes, I was stunned by how steep the demand curve was in China. PRC gold demand was unlike in any other country because, precisely, it was upward sloping – the more expensive the gold, the more the Chinese bought of it. The trend has not changed since then…
Note, how this gold-bug asserts the demand curve for gold is “unlike in any other country because, precisely, it was upward sloping.” This is hardly true, as we can see at least in the anecdotal evidence with demand for gold in the United States — the hysteria for gold increases as the price of the metal increases here as well. This pattern of behavior is not unusual if we assume gold is exhibiting the kind of money-like qualities associated with appreciating currencies. As a currency appreciates, demand for it increases. This suggests that price is driving demand, not vice-versa, that the demand curve for gold is upward sloping — which is to say, the higher the price rises, the greater the demand for gold. Moreover, there is no evidence of a price point, no matter how high, where the demand for gold goes to zero.
To argue this another way: In the real world, economists argue that deflation reduces the willingness of individuals to part with their money for commodities. They hold onto it as they anticipate even lower prices in the future. It is clear that gold is behaving in this fashion — as its price increases — which is to say, as its purchasing power increases — people want to hold onto it, and hold more of it. A hypothesis which does not account for this money-like behavior is not a hypothesis at all.
However, even if there is no price point where the demand for gold goes to zero, this does not mean there is no price point where “flow” goes to zero. If gold does indeed exhibit money-like qualities with an upward sloping ‘demand curve’, this would imply gold can fall to some price below which it no longer circulates as money. We can return to this point later as well.
Krugman now turns to the core question of his post:
So what determines the price of gold at any given point in time? Hotelling models say that people are willing to hold onto an exhaustible resources because they are rewarded with a rising price.
At this point we should say something about this “Hotelling model”. Harold Hotelling developed an economic model to describe how cartels act to restrain the supply of a commodity in the market in order to maximize profit, that is, the return on their investment in the production of the commodity. The Wikipedia has this to say about Hotelling’s Rule regarding scarcity rent — excess profit derived by creating scarcity in the supply of a product:
Hotelling’s rule defines the net price path as a function of time while maximising economic rent in the time of fully extracting a non-renewable natural resource. The maximum rent is also known as Hotelling rent or scarcity rent and is the maximum rent that could be obtained while emptying the stock resource. In an efficient exploitation of a non-renewable and non-augmentable resource, the percentage change in net-price per unit of time should equal the discount rate in order to maximise the present value of the resource capital over the extraction period.
Simply stated, if I have a commodity that will eventually be exhausted, I will manage its production so that, over the lifetime of its production, the amount of money I can charge for it will be maximized. Think about, for instance, OPEC, who wants to be sure they produce no more oil each year than is demanded by the market when the price of oil is the highest and the amount demand is the greatest.
The problem with applying this rule to a stock of gold is that, as we saw above, gold exhibits the characteristic features of a money, not of an ordinary commodity. This will seem to be a non sequitur to Krugman’s core argument — until you realize the aim of maximizing rent on the production of a commodity is to maximize the quantity of money one receives in return for that commodity. Essentially, Krugman is arguing that owners of a lifeless hoard of gold sitting in a vault seek to maximize rent on that lifeless hoard of gold sitting in a vault.
Since the gold never moves from the vault, never enters into circulation, never exchanges with other commodities, and, thereby, become the form of the profits sought by producers of commodities, its role in its own price appreciation or depreciation must be completely passive — it is a mere victim of circumstance, a bystander to events. Whatever the change in the price of gold that occurs must be the result of other processes in the economy that impose themselves on the price of gold, causing this price to vary over time.
So, whatever is happening with the price of gold is not the result of any change in the behavior of the owners of the commodity, nor of any rent maximizing effort on their part. In fact, from what we have seen above, there is no reason to assume the owners of gold do anything with this gold except hold onto it. The entire point of having the gold is to hold it irrespective of any change in its price. While there may be some fluctuations of willingness to hold gold at the margins — of interest in supplies of newly produced gold — the great bulk of gold is likely no more traded than do people trade their savings in any other form. The question raised by this is obvious:
What determines the preference of individuals to hold their savings in the form of gold as opposed to some other form? But, we will leave this to the side as well for now.
Krugman next states:
Abstracting from storage costs, this says that the real price [of gold] must rise at a rate equal to the real rate of interest.
As with the “real price” of gold, I am at a loss at to what the “real rate of interest” refers. So, I went looking for a definition of the term on Wikipedia and found this:
“The nominal interest rate is the amount, in money terms, of interest payable.
For example, suppose a household deposits $100 with a bank for 1 year and they receive interest of $10. At the end of the year their balance is $110. In this case, the nominal interest rate is 10% per annum.
The real interest rate, which measures the purchasing power of interest receipts, is calculated by adjusting the nominal rate charged to take inflation into account. (See real vs. nominal in economics.)
If inflation in the economy has been 10% in the year, then the $110 in the account at the end of the year buys the same amount as the $100 did a year ago. The real interest rate, in this case, is zero.”
Krugman is arguing that the price of gold will rise or fall to reflect interest rates once inflation has been stripped out of the equation. If the real interest rate is positive, gold will tend to appreciate relative to currency. If the real interest rate is negative, gold will tend to depreciate relative to currency. If, at the end of a year $100 in your savings account has increased to $110, and inflation that year is zero, an ounce of gold will appreciate by a proportional amount — say, from $1400 to $1540. If, at the end of a year $100 in your savings account has decreased to $90, and inflation that year is zero, an ounce of gold will depreciate by a proportional amount — say, from $1400 to $1260.
This latter example would likely cause some difficulties: you would go storming into your local bank branch to inquire why you were being charged an astonishing ten percent a year to keep your money in the bank. Once informed that the current interest rate charge by your bank was now -10% per year, you would promptly withdraw your funds — triggering what, in time, will grow into a run on the bank, as everyone withdraws their saving in the face of stiff new negative interest rates.
Why might this cause some difficulties? Between 1980 and 2001, the average annual price of gold fell on average by 5 percent per year; while, since 2001, the average annual price of gold has risen on average by 15 percent per year. The surprising result of Krugman’s argument is that, after accounting for inflation, real interest rates were negative for most of the 80s and 90s, but have been decidedly positive since then.
We will leave this for later examination as well.
Krugman concludes the recent jump in the price of gold is the result of the Federal Reserve Bank’s zero interest rate policy:
Now ask the question, what has changed recently that should affect this equilibrium path? And the answer is obvious: there has been a dramatic plunge in real interest rates, as investors have come to perceive that the Lesser Depression will depress returns on investment for a long time to come:
What effect should a lower real interest rate have on the Hotelling path? The answer is that it should get flatter: investors need less price appreciation to have an incentive to hold gold.
There are two things I question about this reasoning. First, the price of a troy ounce of gold has been increasing since 2001, when it hit bottom at an annual average price of $271. That means, for whatever reason having nothing to do with the Fed’s zero interest rate policy, investors have had an incentive to hold gold as its purchasing power, measured in dollars, has been rising for a decade now. Second, since in my argument, gold is playing only a passive role, the historical evidence suggests the Fed’s zero interest rate policy is being driven by the same forces that are also causing gold to appreciate in price and investors to hoard it.
Rather than driving events, the Fed’s zero interest rate policy is completely reactive. Simply stated, based on Krugman’s argument, the Fed’s zero interest rate policy is not sending capitals scurrying into gold and driving gold price higher, rather it is responding to whatever economic forces are doing this, and, driving real interest rates to an average 15% a year for the last decade — it is trying to drive real interest rates negative to reverse those forces, and to reverse the depressed return on investment.
We will show why this argument falls flat on its face as well
The logic, if you think about it, is pretty intuitive: with lower interest rates, it makes more sense to hoard gold now and push its actual use further into the future, which means higher prices in the short run and the near future.
The evidence is, in fact, the exact opposite: the behavior of gold indicates the Federal Reserve’s zero interest rate policy is a failure so far (along with all the fiscal stimulus and backdoor bailouts) since, despite the effort and unprecedented scale of the various policy actions, the price of gold indicates interest rates remain stubbornly high at levels not seen since the 1970s depression. And, moreover, still increasing.
Nevertheless his string of errors in reasoning, Krugman manages to end up, Mr. Magoo-like, at what is somewhat close to the right conclusion:
…this is essentially a “real” story about gold, in which the price has risen because expected returns on other investments have fallen; it is not, repeat not, a story about inflation expectations. Not only are surging gold prices not a sign of severe inflation just around the corner, they’re actually the result of a persistently depressed economy stuck in a liquidity trap — an economy that basically faces the threat of Japanese-style deflation, not Weimar-style inflation. So people who bought gold because they believed that inflation was around the corner were right for the wrong reasons.
Krugman is correct to state rising gold price is a sign of an economy in a depression, where returns on investment have fallen flat. He is also correct to state gold is not signalling future inflation. But, Krugman arrives the correct conclusion only by making a series of Mr. Magoo-like blunders that just manage to offset each other — blunders, which, when stripped out of his argument, allow a simpler explanation for the relation between gold and real interest rates.
In the next part of this series, I will show why Krugman’s model, although arriving at something close to the truth of the matter, is nevertheless wholly wrong.